Mostra el registre d'ítem simple

dc.contributor.authorRadojkovic, Petar
dc.contributor.authorCakarevic, Vladimir
dc.contributor.authorVerdú Mulà, Javier
dc.contributor.authorPajuelo González, Manuel Alejandro
dc.contributor.authorCazorla Almeida, Francisco Javier
dc.contributor.authorNemirovsky, Mario
dc.contributor.authorValero Cortés, Mateo
dc.contributor.otherUniversitat Politècnica de Catalunya. Departament d'Arquitectura de Computadors
dc.date.accessioned2010-06-29T10:57:54Z
dc.date.available2010-06-29T10:57:54Z
dc.date.created2010-01
dc.date.issued2010-01
dc.identifier.citationRadojkovic, P. [et al.]. Thread to strand binding of parallel network applications in massive multi-threaded systems. A: 15th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming. "Proceedings of the 15th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming". Bangalore, India: ACM Press. Association for Computing Machinery, 2010, p. 191-202.
dc.identifier.isbn978-1-60558-708-0
dc.identifier.urihttp://hdl.handle.net/2117/7885
dc.description.abstractIn processors with several levels of hardware resource sharing, like CMPs in which each core is an SMT, the scheduling process becomes more complex than in processors with a single level of resource sharing, such as pure-SMT or pure-CMP processors. Once the operating system selects the set of applications to simultaneously schedule on the processor (workload), each application/thread must be assigned to one of the hardware contexts (strands). We call this last scheduling step the Thread to Strand Binding or TSB. In this paper, we show that the TSB impact on the performance of processors with several levels of shared resources is high. We measure a variation of up to 59% between different TSBs of real multithreaded network applications running on the UltraSPARC T2 processor which has three levels of resource sharing. In our view, this problem is going to be more acute in future multithreaded architectures comprising more cores, more contexts per core, and more levels of resource sharing. We propose a resource-sharing aware TSB algorithm (TSBSched) that significantly facilitates the problem of thread to strand binding for software-pipelined applications, representative of multithreaded network applications. Our systematic approach encapsulates both, the characteristics of multithreaded processors under the study and the structure of the software pipelined applications. Once calibrated for a given processor architecture, our proposal does not require hardware knowledge on the side of the programmer, nor extensive profiling of the application. We validate our algorithm on the UltraSPARC T2 processor running a set of real multithreaded network applications on which we report improvements of up to 46% compared to the current state-of-the-art dynamic schedulers.
dc.format.extent11 p.
dc.language.isoeng
dc.publisherACM Press. Association for Computing Machinery
dc.subjectÀrees temàtiques de la UPC::Informàtica::Arquitectura de computadors
dc.subject.lcshComputer Systems Performance
dc.subject.lcshThreads (Computer programs)
dc.subject.lcshComputer simulation -- Software
dc.titleThread to strand binding of parallel network applications in massive multi-threaded systems
dc.typeConference report
dc.contributor.groupUniversitat Politècnica de Catalunya. CAP - Grup de Computació d'Altes Prestacions
dc.identifier.doi10.1145/1693453.1693480
dc.description.peerreviewedPeer Reviewed
dc.relation.publisherversionhttp://dl.acm.org/citation.cfm?doid=1693453.1693480
dc.rights.accessRestricted access - publisher's policy
drac.iddocument2568291
dc.description.versionPostprint (published version)
upcommons.citation.authorRadojkovic, P.; Cakarevic, V.; Verdu, J.; Pajuelo, A.; Cazorla, F.; Nemirovsky, M.; Valero, M.
upcommons.citation.contributorACM SIGPLAN Symposium on Principles and Practice of Parallel Programming
upcommons.citation.pubplaceBangalore, India
upcommons.citation.publishedtrue
upcommons.citation.publicationName15th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming
upcommons.citation.startingPage191
upcommons.citation.endingPage201


Fitxers d'aquest items

Imatge en miniatura

Aquest ítem apareix a les col·leccions següents

Mostra el registre d'ítem simple

Tots els drets reservats. Aquesta obra està protegida pels drets de propietat intel·lectual i industrial corresponents. Sense perjudici de les exempcions legals existents, queda prohibida la seva reproducció, distribució, comunicació pública o transformació sense l'autorització del titular dels drets